Per discussions in #76 and #77, we should just remove all the cron stuff.
It's overly complex, has lots of problems (breaks on large sites, especially on initial activation) and doesn't match our use cases. Instead, we should document the CLI pieces better and provide examples on how best to configure ongoing updates.
Thinking this through a bit more (but open to discussion):
For sites that don't have access to a queue/tasks/jobs system, it would be impossible to run non-blocking updates for sitemaps (e.g. after a post is updated). So, we may want to keep cron around for individual sitemap updates, but remove the initial cron build and the generate/regenerate buttons.
Hello, any update on this?
I have a large site where the cron process takes a very long time, slows down the site and seems to cause problems with scheduling posts and other cron-based tasks.
I think disabling all cron tasks by default and being able to use wp-cli or a special URL to call via curl (this one because services such as Pantheon don't allow running wp-cli commands from an external cron job) would be very helpful.
Kill full sitemap cron generation
For really large datasets, this ends up in a flood of thousands of cron events that can
slow down or completely disable a site.